Proposed Experiment

[Proposed Experiment] Problem for Financial Institutions Fiji July 2019

Lean Phase: Problem

Assumption: Financial Institutions have trouble accessing certain customers, or adequately profiling existing customers, due to factors such as

  • Poor information on the existing customer
  • Poor risk modelling methods
  • Inability or lack of resources to search for new customers  
  • Customers not understanding what is needed to access a solution from them

Time Box: 2 weeks

Success Metric:

Whether the proposed customer segment identifies with the assumed problem, measured by the % of FIs that seeking a solution to the assumed problem.

Team will interview 8 FIs.


Green Light: Proceed to offer testing the UVP.

Success Point: 50% of FIs identify with the problem.

Orange Light: Re-assess the problem definition based on research gathered during the experiment and redefine the next experiment around the assumed new ‘problem’.

Failure Point: Less than 30% of FIs identify with the problem.

Red Light: Assess if any interviews indicate a problem that can be defined, perhaps outside of the space we have looked at. Re-assess customer segment and problem assumptions.

Experiment Build

  1. Create an interview script and survey to carry out this experiment

  2. Organise to meet 8 FIs to discuss this issue. [Make sure to consult February experiment results on interviewing FIs, and consult HubSpot as to which FIs have previously been contacted by PEV teams.]

  3. Conduct rehearsals and training with the team to prepare them for interviews.

  4. Conduct 8 interviews with FIs to define the problem

  5. Post-interview: collate data and create problem definition

Questions to ask

Define the institution

What kind of institution are you? Bank, MFI, formal/informal institution?

What financial/business solutions do you offer? - Loans/credit, financial advice, budgeting/savings systems, etc.?

  • Gather information on what kinds of loans they give, how much, to whom, etc. (Lots of this information is gathered by February teams, but we will be doing this survey with more FIs than they interviewed, so these questions are still necessary and to validate February's data).
  • What information do you require from a customer to access these solutions? What is a legal requirement and what do you put in place? (We have information on this for loans from Feb, but need to understand what is a legal requirement and not).

What issues do you have when assessing customers for these solutions

What systems do you currently use to model credit risk? Is this effective? What are your default rates?

What systems do you currently have for gathering information to model this risk? Is this effective? What information do you require from customers? Is it easy to acquire this information? What do you consider 'good' and 'bad' customers, and how would you turn a 'bad' customer into a 'good' customer?

How do you get proof of cash flow? What would you need from a customer in order to consider them having 'good' and 'bad' cash flow? 

What are the reasons you don’t take on certain customers? What would make you comfortable taking on these new customers? What makes a good customer and a bad customer?

How much does it cost to assess someone for a financial product? How much time does it take? Do you use third-parties to do this, or is it purely internal processes?

edited on 21st May 2019, 06:05 by Rose Martin

Rose Martin 1 month ago

Status label added: Proposed Experiment

Reply 0